The experiments in this proposal are designed to increase understanding of the mechanisms by which DNA is repaired and recombined in eukaryotic cells. The specific goals are concerned with pathways of the DNA mismatch repair system, which is responsible for the correction of non- complementary bases in the DNA helix. In prokaryotes this pathway is known to be important in correcting replication errors, deaminations of 5-methylcytosine, and heteroduplexes in DNA which have arisen by homologous recombination. the prokaryotic mismatch repair system also prevents recombination between homologous (similar but not identical) sequences and its absence leads to large chromosomal duplications. MutS is the protein responsible for recognizing mismatches in E. coli. In wild-type strains, MutS suppresses recombination of UV-damaged DNA, but in excision repair-deficient strains, MutS actually increases recombination of UV-damaged DNA. There are also indications that mismatch repair in prokaryotes is involved in repairing potential mutations created by the UV-induced repair pathway and in repairing abnormal or damaged DNA that arises due to base analog incorporation, or damage by some alkylating agents and mutagens. It remains to be seen whether any human diseases are due to defects in the DNA mismatch repair pathway, but increased spontaneous mutation rates could give rise to increased incidences of tumors and other diseases, and failure of the pathway could lead to a greater susceptibility to the deleterious effects of radiation, mutagens, and other damaging agents. In addition, the effect of the mismatch repair system on recombination could be extremely important. In normal human cells, the mismatch repair pathway is presumably involved in preventing inappropriate recombination. However, two different xeroderma pigmentosum (XP) complementation groups show an increase in UV-stimulated recombination. This increase in recombination after UV damage could be an important component of the increased risk of certain malignancies. Compared to the wealth of information known about mismatch repair in prokaryotes, very little is known about mismatch repair in eukaryotes. Clones of a homologue of the mutS gene have been obtained in both yeast (MSH3) and mouse (Rep-3) cells. Evidence to date suggests that these genes have a primary role in prevention of homologous recombination rather than in mutation avoidance. Our specific aims are: (1) to study the function of the yeast MSH3 gene, to determine its role in homologous recombination, and to determine to what substrates it binds and the other proteins it interacts with; (2) to understand the functioning of the mouse Rep-3 gene, whether it plays a role on homologous recombination in mouse cells, and to what substrates it binds; and (3) to isolate additional mouse MutS homologues and analyze their function in DNA repair and recombination.